摘要
针对车辆电源系统测试点少且测试数据不完备的问题,提出一种多信号流图模型和贝叶斯网络相结合的故障诊断方法。利用多信号流图模型建立电源系统的故障诊断模型,得到系统故障源与测试信号对应的故障依赖矩阵,在此基础上,建立用于故障诊断的贝叶斯网络结构,根据历史数据完成网络的参数学习,并以故障后验概率最大为准则,实现电源系统的故障诊断。仿真实验验证了该方法的有效性。
The vehicle power system has the fewer test points and the testing data are incomplete.Aiming at these characteristics,it proposes that combining multi-signal flow graph model with Bayesian network fault diagnosis method.The fault diagnosis model of power system is built by applying multi-signal flow graph.The dependence matrix which relates faults and testing signals is generated based on the model,and setting up the corresponding Bayesian network structure for the fault diagnosis,based on historical data to complete the network parameter learning.Using the maximum posterior probability of failure as a criterion to achieve the fault diagnosis of power system.Simulation results verify the effectiveness of the method.
出处
《计算机工程》
CAS
CSCD
北大核心
2011年第23期251-253,共3页
Computer Engineering
关键词
电源系统
多信号流图模型
贝叶斯网络
故障诊断
参数学习
power system
multi-signal flow graph model
Bayesian network
fault diagnosis
parameter learning